Auto IV: Counterfactual Prediction via Automatic Instrumental Variable Decomposition

@article{Yuan2022AutoIC,
  title={Auto IV: Counterfactual Prediction via Automatic Instrumental Variable Decomposition},
  author={Junkun Yuan and Anpeng Wu and Kun Kuang and B. Li and Runze Wu and Fei Wu and Lanfen Lin},
  journal={ACM Trans. Knowl. Discov. Data},
  year={2022},
  volume={16},
  pages={74:1-74:20}
}
Instrumental variables (IVs), sources of treatment randomization that are conditionally independent of the outcome, play an important role in causal inference with unobserved confounders. However, the existing IV-based counterfactual prediction methods need well-predefined IVs, while itโ€™s an art rather than science to find valid IVs in many real-world scenes. Moreover, the predefined hand-made IVs could be weak or erroneous by violating the conditions of valid IVs. These thorny facts hinder theโ€ฆย 

Figures and Tables from this paper

Machine Learning in Causal Inference: Application in Pharmacovigilance
TLDR
A literature search revealed that the adoption of causal paradigms can mitigate known issues with machine learning models, and it is predicted that the pharmacovigilance domain can benefit from the progress in the machine learning field.
A Model-Agnostic Causal Learning Framework for Recommendation using Search Data
TLDR
A model-agnostic framework named IV4Rec is proposed that can effectively decompose the embedding vectors into these two parts, hence enhancing recommendation results and demonstrates that it consistently enhances RSs and outperforms a framework that jointly considers search and recommendation.

References

SHOWING 1-10 OF 60 REFERENCES
Valid Causal Inference with (Some) Invalid Instruments
TLDR
This paper shows how to perform consistent IV estimation despite violations of the exclusion assumption, and shows that when one has multiple candidate instruments, only a majority of these candidates---or, more generally, the modal candidate-response relationship---needs to be valid to estimate the causal effect.
Learning Decomposed Representation for Counterfactual Inference
TLDR
The proposed synergistic learning framework can precisely identify and balance confounders, while the estimation of the treatment effect performs better than the state-of-the-art methods on both synthetic and real-world datasets.
Deep IV: A Flexible Approach for Counterfactual Prediction
TLDR
This paper provides a recipe for augmenting deep learning methods to accurately characterize causal relationships in the presence of instrument variables (IVs)โ€”sources of treatment randomization that are conditionally independent from the outcomes.
Ivy: Instrumental Variable Synthesis for Causal Inference
TLDR
Ivy is a new method to combine IV candidates that can handle correlated and invalid IV candidates in a robust manner and can correctly identify the directionality of known relationships and is robust against false discovery.
Learning Disentangled Representations for CounterFactual Regression
TLDR
This work proposes an algorithm to identify disentangled representations of the above-mentioned underlying factors from any given observational dataset D and leverage this knowledge to reduce, as well as account for, the negative impact of selection bias on estimating the treatment effects from D.
Kernel Instrumental Variable Regression
TLDR
KIV is proposed, a nonparametric generalization of 2SLS, modeling relations among X, Y, and Z as nonlinear functions in reproducing kernel Hilbert spaces (RKHSs), and it is proved the consistency of KIV under mild assumptions, and conditions under which convergence occurs at the minimax optimal rate for unconfounded, single-stage RKHS regression.
Counterfactual VQA: A Cause-Effect Look at Language Bias
TLDR
A novel counterfactual inference framework is proposed, which enables the language bias to be captured as the direct causal effect of questions on answers and reduced by subtracting the direct language effect from the total causal effect.
CausalVAE: Disentangled Representation Learning via Neural Structural Causal Models
TLDR
This work proposes a new VAE based framework named CausalVAE, which includes a Causal Layer to transform independent exogenous factors into causal endogenous ones that correspond to causally related concepts in data.
On the Use of the Lasso for Instrumental Variables Estimation with Some Invalid Instruments
TLDR
This work investigates the behavior of the Lasso for selecting invalid instruments in linear instrumental variables models for estimating causal effects of exposures on outcomes, and proposes a median estimator that is consistent when less than 50% of the instruments are invalid.
Treatment Effect Estimation via Differentiated Confounder Balancing and Regression
TLDR
A data-driven Differentiated Confounder Balancing algorithm to jointly select confounders, differentiate weights of confoundingers and balance confounder distributions for treatment effect estimation in the wild high-dimensional settings is proposed and shows that the top features ranked by the algorithm generate accurate prediction of online advertising effect.
...
...